Proton‐Exchanged Montmorillonite With Interlayer Accessibility as a Solid Acid Catalyst for the Upcycling of Polyoxymethylene

ABSTRACT

Engineering plastics such as polyoxymethylene (POM) are highly resistant to depolymerization due to their robust C─O backbones, posing a major challenge for sustainable recycling. Herein, we report a solid acid catalyst based on proton‐exchanged montmorillonite (H‐Mont) that efficiently upcycles POM into cyclic acetals under mild conditions. H‐Mont exhibits high catalytic performance, converting both pristine and post‐consumer POM, outperforming conventional solid and liquid acid catalysts in both catalytic efficiency and reusability. Mechanistic studies reveal that interlayer accessibility generated by solvent‐induced swelling, together with strong Brønsted acidity of H‐Mont, synergistically overcomes interfacial limitations between the catalyst and semi‐crystalline POM, enabling efficient depolymerization. This represents the first demonstration of a streamlined, reusable heterogeneous catalysis for POM upcycling toward cyclic acetals and establishes a strategy for transforming chemically resilient engineering plastics into valuable chemicals.

The first microwave‐assistance‐free heterogeneous catalytic upcycling of polyoxymethylene (POM) into valuable cyclic acetals is achieved with a reusable proton‐exchanged montmorillonite (H‐Mont) catalyst. Combining interlayer accessibility with strong acidity, H‐Mont enables efficient, regenerable, and versatile transformations with potentially biomass‐derived diols, advancing sustainable strategies for the valorization of engineering plastics.

1. Introduction

Engineering plastics, prized for thermal and chemical durability, represent one of the most formidable challenges to chemical recycling [1, 2, 3, 4, 5]. Among them, polyoxymethylene (POM), a semi‐crystalline thermoplastic widely used in automotive and consumer applications, is particularly resistant to depolymerization due to its stable C─O backbone [6, 7, 8, 9, 10, 11, 12]. This intrinsic robustness severely limits its chemical valorization under mild conditions.

Recent studies have explored the upcycling of POM using liquid acids and metal complexes, enabling its depolymerization into formaldehyde, which can serve as a reactive C₁ synthon for diverse downstream transformations, such as acetalization, methanol synthesis, and C─C/C─N bond‐forming reactions [8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20]. While these systems have advanced mechanistic understanding and enabled precise catalytic transformations, most of them rely on homogeneous catalysts, presenting major challenges for process implementation, including difficulties in catalyst separation, limited reusability, and poor scalability [3, 7, 10, 13, 14, 21, 22]. Thus, there remains a strong need for practical and reusable heterogeneous catalysts capable of directly upcycling POM under mild conditions.

However, the application of solid catalysts to semi‐crystalline polymers such as POM remains underdeveloped, largely because insufficient solid–solid contact limits polymer accessibility and suppresses catalytic activity [21, 23, 24, 25, 26, 27]. Current heterogeneous approaches, such as oxidation with 1,1,1,3,3,3‐hexafluoroisopropanol (HFIP)‐assisted dissolution [11] still rely on cost‐prohibitive and industrially impractical fluorinated solvents, while even recent routes, either by using heterogeneous resins to produce heterocycles [27] or zeolite‐based upcycling to methanol [20] require microwave irradiation or high temperatures (≥ 200°C) respectively, limiting their feasibility for scalable POM upcycling.

To overcome these limitations, we devised a solid catalyst platform that integrates enhanced polymer accessibility with tunable acidity. Montmorillonite (Mont), a layered aluminosilicate with exchangeable interlayer Na⁺ ions, was selected for two reasons: (1) its high swelling capacity in polar media, which enhances contact between polymer chains and catalytic sites [28, 29, 30, 31, 32]; and (2) its high cation‐exchange capacity, allowing precise control over acid strength [28, 29, 32]. These structural and chemical characteristics provide a rational basis for catalyst design tailored to the demands of semi‐crystalline polymer depolymerization.

In this study, we report a proton‐exchanged Mont (H‐Mont) catalyst that enables the efficient upcycling of both pristine and consumer‐product‐derived POM waste into cyclic acetals under mild conditions. H‐Mont significantly outperforms both conventional homogeneous and heterogeneous acid catalysts, representing the first example of a reusable solid catalyst system for POM upcycling toward cyclic acetals without the addition of HFIP or microwave assistance. This work establishes a rational catalyst design strategy that enables the efficient and easily scalable upcycling of depolymerization‐resistant engineering plastics such as POM under mild conditions without the need for inaccessible solubilizing agents or microwave irradiation.

2. Results and Discussion

To evaluate the catalyst design strategy that integrates polymer accessibility and acid strength modulation in Mont, we systematically investigated the catalytic performance of various ion‐exchanged Mont catalysts, along with several referential solid acid catalysts, for the upcycling of POM into 1,3‐dioxane with 1,3‐propanediol in 1,4‐dioxane solvent at 130°C for 10 min (Figure 2a). For optimization details, refer to Tables S1 and S2.

FIGURE 2.

Catalytic performance of H‐Mont (a) Catalyst screening for POM upcycling toward 1,3‐dioxane using various solid acid catalysts. Reaction conditions: 50 mg of POM, 50 mg of catalyst, 2 mmol of 1,3‐propanediol at 130°C in 2 mL of 1,4‐dioxane for 10 min. The yield was determined by ¹H‐NMR using an internal standard. *The reaction time was 20 min. **The catalyst loading was reduced to 12.5 mg of H‐Mont for 30 min. ***0.083 mmol of 98% H₂SO₄ was used. (b) H‐Mont reuse capabilities in the upcycling toward 1,3‐dioxane. Reaction conditions: 50 mg of POM, 50 mg of H‐Mont, 2 mmol of 1,3‐propanediol at 130°C in 2 mL of 1,4‐dioxane for 30 min, the polymer‐to‐catalyst mass ratio was kept constant by reducing the POM mass upcycled to equate the original ratio, overcoming any possible catalyst mass loss between cycles. After the seventh run, the spent catalyst was regenerated by HCl treatment. (c) Applicability of the H‐Mont system with POM waste from diverse sources at low catalyst loading. Reaction conditions are described in the reaction scheme shown. (d) H‐Mont scale‐up capabilities in the upcycling toward 1,3‐dioxane. (i) POM waste was obtained from laboratory clips made of Delrin. (ii) 10 g of pristine POM pellets (Duracon M90‐44).

2.1. Catalyst Screening

Among the Mont catalysts tested, H‐Mont exhibited the highest activity, affording 1,3‐dioxane in 85% yield. Other cation‐exchanged Monts, such as Sn‐Mont, Ti‐Mont, and Sc‐Mont, demonstrated moderate activity, while the parent Na‐Mont showed no activity, highlighting the importance of the interlayer cation in governing catalytic performance. Mont K10 also showed activity comparable to Ti‐Mont (52% and 54%, respectively), but lower than H‐Mont's. All evaluations were conducted in 1,4‐dioxane, which was selected due to its low cost and practical availability. Preliminary solvent screening revealed that HFIP, as a highly polar protic solvent, also afforded a high yield (around 99%) under identical conditions (Table S3), presumably owing to enhanced solvation and improved catalyst swelling.

Notably, H‐Mont significantly outperformed a series of conventional solid acid catalysts, including zeolites (H‐Mordenite, H‐ZSM‐5, H‐Beta Zeolite – 25, H‐Beta Zeolite – 150, USY‐Zeolite 10, and USY‐Zeolite 40), ion‐exchange resins (Amberlyst‐36), perfluorosulfonic acid polymers (Nafion), and even the widely studied solid superacid SO₄•ZrO₂ [33, 34], all of which exhibited markedly low to no activity under identical conditions. For comparison, sulfuric acid (H₂SO₄), a representative homogeneous Brønsted acid widely used in related reactions [16, 19, 35, 36, 37], provided a 50% yield under the same conditions, indicating that H‐Mont is approximately twice as effective. Furthermore, when the reaction time was extended to 30 min, H‐Mont afforded a quantitative yield (96%) of 1,3‐dioxane, highlighting its outstanding efficiency even under mild conditions and low catalyst loading (97% with H‐Mont loading of 25% wt). Taken together, these results demonstrate H‐Mont as a highly effective and structurally adaptive solid acid catalyst for the upcycling of depolymerization‐resistant polymers such as POM.

2.2. Reusability

The durability of H‐Mont was next assessed under standard reaction conditions (Figure 2b). After each cycle, the catalyst was readily recovered by simple centrifugation without any specialized treatment and reused in subsequent runs. H‐Mont maintained high catalytic activity over six consecutive cycles, consistently affording yields above 94%. (Even under low catalyst loading conditions (25% wt), three successful consecutive cycles afforded yields over 87% (Figure S1). A sudden decline in activity was observed between the sixth and seventh runs, which may be attributed to surface deactivation caused by the accumulation of organic species or partial loss of acid sites (Table S4 and Figure S2). Post‐reaction characterization (Figures S3 and S4) confirmed this interpretation, showing reversible blockage of Brønsted and Lewis acid sites by organic residues. Hot filtration experiments showed that no active sites are leached during reaction, showing the stability of H‐Mont (Figures S5 and S6). Importantly, the spent catalyst could be regenerated by washing with hot solvent followed by acid treatment analogous to the original preparation, fully restoring its activity in the eighth and ninth runs. These results demonstrate not only the practical reusability of H‐Mont but also its operational robustness, highlighting its potential as a durable heterogeneous catalyst for POM upcycling.

2.3. Catalytic Applicability Studies

The versatility of the H‐Mont catalyst extends to a broad range of diol nucleophiles (Table S5). Acetalization of three primary diols: ethylene glycol, 1,4‐butanediol, and 1,5‐pentanediol, each of which can be synthesized by promising biomass‐derived routes [38, 39, 40]; afforded 1,3‐dioxolane, 1,3‐dioxepane, and 1,3‐dioxocane in 82%, 63%, and 60% yields, respectively (entries 1, 3, and 4). Yields depend strongly on diol structure: the secondary diol 2,4‐pentanediol gave 4,6‐dimethyl‐1,3‐dioxane in much lower yield than the 4‐methyl‐1,3‐dioxane obtained from 1,3‐butanediol, which contains one unhindered primary hydroxyl group (entry 6 vs 5). This contrast underscores the importance of a primary hydroxyl group and the possibility of steric hindrance around the secondary ‐OH that slows ring closure. Upcycling with all these diols exceeds or falls within close range of the best yields reported for acid catalysts (Table S5) [13, 14, 15, 20], highlighting the superior efficiency and robustness of the present H‐Mont system (characterization of the crude reactive media is presented in Figures S7–S24).

To demonstrate the practical applicability of this catalytic system, we investigated the upcycling of post‐consumer POM products. A range of POM‐containing commercial items, including laboratory joint clips, door latches, correction fluid caps, mechanical gears, and other molded plastic parts, were manually chipped and subjected to the standard reaction conditions without any chemical pre‐treatment (Figure 2c).

Despite the rigid and insoluble nature of these materials, which pose substantial barriers to catalytic processing, H‐Mont efficiently converted all samples into the corresponding cyclic acetal products in high yields ranging from 79% to 93%. Notably, the upcycling of laboratory joint clips, a representative example of chemically resilient and geometrically complex POM waste, afforded a 93% yield of 1,3‐dioxane, underscoring the system's remarkable compatibility with real‐waste substrates.

To further assess the scalability of the system, the upcycling of chipped laboratory clips and pristine POM plastic pellets was conducted on preparative scales using 1 g and 10 g of each sample, respectively. These scale‐up experiments afforded 1,3‐dioxane in 77% and 86% yields, respectively (Figure 2d), confirming that the proposed catalytic platform is not only versatile in terms of substrate scope but also operationally viable for the conversion of real‐world plastic waste into value‐added chemicals.

2.4. Properties‐Activity Correlation

To elucidate the origin of H‐Mont's superior catalytic performance, we investigated the relationship between its structural characteristics and catalytic activity. Particular attention was given to two key parameters: interlayer swelling capacity and Brønsted acidity, both of which, resulting from successful ion exchange (Table S6), are expected to influence polymer accessibility and catalytic turnover [41, 42].

Nitrogen adsorption–desorption isotherm analysis revealed that the mesopore volume (V _meso) of H‐Mont was 0.127 cm³g⁻¹, notably larger than that of H‐ZSM‐5 (0.058 cm³g⁻¹), a representative conventional solid acid (Figure S25 and Table S7) [32, 33]. This result suggests that H‐Mont possesses a more open pore structure favorable for interaction with polymer substrates [43, 44, 45].

The swelling behavior of H‐Mont was examined by XRD (Figure 3a). Treatment with 1,4‐dioxane shifted the (001) reflection toward lower angles, corresponding to an expansion of the interlayer space from 3.0 Å to 5.5 Å. The other montmorillonite samples showed a similar tendency, with the exception of Sn‐Mont, which does not have a layered structure (Figure S26) [29, 46, 47]. Upon addition of POM, the interlayer spacing increased slightly further to 5.7 Å, consistent with POM accessing the swollen galleries of H‐Mont and thereby reaching the catalytically active sites.

FIGURE 3.

Structure–activity relationship of H‐Mont (a) Normalized XRD patterns showing the (001) reflection of H‐Mont under different treatments. The diffractograms exhibit a shift of the (001) reflection toward lower angles, indicating an increase in interlayer spacing within the montmorillonite structure. Interlayer space was calculated using Bragg's law, subtracting the intrinsic montmorillonite layer thickness (9.6 Å). (b) Normalized NH₃‐TPD profiles of various solid acid catalysts to compare acid strength; the desorption peaks are marked with dotted lines. (c) Reaction progression analyzed by GPC. Reaction conditions: 100 mg POM, 12.5 mg H‐Mont, 4 mmol 1,3‐propanediol at 130°C in 2 mL of 1,4‐dioxane. The peaks at 18 min correspond to the cyclic portions of POM. (d) Information for linear polymer residues during reaction progression experiments, obtained by GPC, compared to reaction yields. *Calculated by GPC using PMMA calibration curve and obtaining the results as PMMA equivalents. (e) POM upcycling mechanism with 1,3‐propanediol towards 1,3‐dioxane aided by H‐Mont solid catalysts.

The acidity of H‐Mont was evaluated by ammonia temperature‐programmed desorption (NH₃‐TPD), together with reference catalysts (Figures 3b and S27). H‐Mont exhibited a pronounced feature above 500°C, characteristic of strong acid sites [45, 48, 49], whereas H‐ZSM‐5, with lower activity, released ammonia predominantly below 500°C. This correlation highlights the critical role of strong acid sites in promoting POM upcycling. Pyridine‐adsorbed FTIR measurements further confirmed that these Brønsted sites persist during catalysis and can be regenerated after recycling, consistent with the reversible surface deactivation observed in the ATR‐FTIR and TPD analyses.

Semi‐quantitative correlation between the temperature of the highest‐temperature NH₃ desorption peak and the catalytic activity of the materials suggests that catalysts with higher‐temperature desorption peaks generally exhibit higher POM depolymerization activity (Figure S28). However, this alone is not enough, as conventional solid acids, such as SO₄•ZrO_2, possessing strong acid sites and regarded as a superacid solid with strong Brønsted and Lewis acidity [33, 34], showed poor activity under identical conditions (Figure 1a). These findings indicate that strong acidity alone is insufficient to drive the reaction. Rather, both strong Brønsted acidity and structural attributes that enable effective substrate access for both diol and POM chains, such as the interlayer swelling property, are required to achieve high catalytic performance.

FIGURE 1.

Conceptual illustration of the first heterogeneous catalytic platform without the need for inaccessible solubilizing agents or microwave irradiation, enabling direct POM upcycling toward cyclic acetals.

2.5. Upcycling Mechanism

To gain further insight into the decomposition behavior of POM, a range of formaldehyde sources and POM samples with different molecular weights were tested, showing that polymer scission is rate‐determining in this system (Table S8). Control experiments conducted in the absence of diol showed that H‐Mont promoted the formation of formaldehyde, trioxane, and several oxygenated intermediates, in agreement with the proposed mechanism (Figure S29). In addition, a time‐resolved two‐stage protocol was used to quantify free formaldehyde generated during the depolymerization of POM and PFA, both formaldehyde‐based polymers with markedly different molecular weights (68 kDa and 0.24–3.0 kDa, respectively), and to monitor its decrease upon addition of diols (for details, see the Supporting Information). H‐Mont uniquely enabled rapid upcycling of polymeric POM, whereas Sc‐Mont, a layered material with lower Brønsted acidity, and SO₄•ZrO₂, a strongly acidic but non‐layered material, showed much lower activity toward POM under identical conditions. The corresponding time‐course data are shown in Figures S30–S32, and the ¹H‐NMR spectra of selected crude reaction mixtures are provided in Figures S33–S44.

Gel permeation chromatography (GPC) was performed (Figure 3c,d). At the early stage of the reaction, two major peaks were observed at retention times of 12–17 and 18–19 min, corresponding to linear and cyclic polymers, respectively. As the reaction proceeded, these peaks gradually broadened and decreased in intensity, indicating the formation of polymer fragments with broadening molecular‐weight distributions. This observation suggests that the depolymerization of POM occurs via random scission along the polymer backbone. The evolution of molecular‐weight distribution further supports this interpretation. As summarized in Figure 3d, both weight‐average molecular weight (Mw) and number‐average molecular weight (Mn) decreased with time, and the dispersity index (Mw/Mn) steadily declined, consistent with random cleavage events. Notably, the decrease in GPC peak area (relative area) closely matched the increase in the yield of cyclic acetals, suggesting that

 acetalization occurs immediately after chain cleavage. These findings collectively offer key mechanistic insight into the catalytic transformation of POM into cyclic acetals under the present conditions.

Based on the structural characterization data, the intermediate visualization, and GPC analyses described above, a plausible reaction mechanism is proposed (Figure 3e). Under the reaction conditions, POM penetrates the expanded interlayer galleries of H‐Mont, where it becomes accessible to strong Brønsted acid sites. Within this confined acidic environment, POM undergoes proton‐catalyzed depolymerization to generate formaldehyde as a reactive intermediate. Given the high reactivity of FA and its spatial proximity to the acid sites, acetalization with 1,3‐propanediol proceeds rapidly, affording 1,3‐dioxane as the final product, with water as the only byproduct (Figure S45).

The exceptional performance of H‐Mont can thus be attributed to the synergistic integration of two key features: interlayer accessibility enabled by structural swelling, together with strong acidity that promotes both efficient chain scission and subsequent acetal formation. This dual functionality underlies the high activity of H‐Mont and highlights its potential as a solid catalyst platform for the upcycling of chemically robust engineering plastics such as POM.

3. Conclusion

This study provides a design framework for the rational design of solid acid catalysts capable of overcoming interfacial and stability limitations in polymer valorization. We demonstrate this through a heterogeneous catalytic system based on H‐Mont, which enables the efficient upcycling of POM into cyclic acetals under mild conditions. H‐Mont significantly outperforms both homogeneous and conventional solid acid catalysts, and uniquely combines strong Brønsted acidity with interlayer swelling capacity to facilitate polymer activation. The catalyst exhibits broad substrate scope and enables direct transformation of post‐consumer POM waste without pretreatment, even at gram scales. It also shows excellent reusability and can be regenerated by simple acid treatment. Together, these findings position H‐Mont as a sustainable and practically viable platform for tackling longstanding challenges in the catalytic valorization of chemically resilient polymers.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

The authors have cited additional references within the Supporting Information [50, 51, 52, 53, 54].

Supporting File: chem71054‐sup‐0001‐SuppMat.docx.